Multiple seed dispersal modes of an invasive plant species on coastal dunes

1 This is the peer-reviewed version of the article accepted for publication in BIOLOGICAL INVASIONS 23:111– 1271. 2021, which has been published in final form at https://doi.org/10.1007/s10530-020-02359-6 Multiple seed dispersal modes of an invasive plant species on coastal dunes Juan B. Gallego-Fernández 1*M. Luisa Martínez 2, José G. García-Franco 2, María Zunzunegui 1 1 Departamento de Biología Vegetal y Ecología, Universidad de Sevilla, Sevilla, Spain 2 Instituto de Ecología, A.C., Xalapa, Veracruz, Mexico * Author for correspondence: [email protected] Abstract Beach evening-primrose (Oenothera drummondii subsp. drummondii) is a perennial herb native to the coastal dunes of the Gulf of Mexico and southeastern USA. During the last century, the species was unintentionally introduced into coastal dune systems around the world. The purpose of this study was to explore the means of dispersal of this invasive in the newly established populations on the coastal dunes of Spain and to determine if they contribute significantly to the invasion process. After reconstructing the history of its spread in Spain from available records, we tested whether dispersion by marine currents can be a mechanism of colonization at regional scale, and whether the presence of native vertebrate participate in the expansion process at local scale. Through laboratory experiments, we demonstrated that the seeds of O. drummondii can be dispersed by seawater, since 0.63% present both buoyancy in seawater and subsequent germination after being washed with fresh water. This capacity for dispersal by marine currents could explain both the expansion of the species along the southwestern coast of Spain and the maintenance of its populations following disturbances such as storms. We also demonstrated that Iberian Hares and European Rabbits can disperse the seeds by endozoochory; seeds recovered from faecal pellets of these species were capable of germination (73 and 63%, respectively). O. drummondii has become an invader only in coastal dunes where these dispersers are present, and with no control actions to regulate their populations. In conclusion, O. drummondii is an alien species with varying dispersal methods, polychory. This helps maintain already established populations and disperses seeds at a regional level, depending on the incidence of storms and the presence or absence of rabbits and hares for efficient local seed dispersion. Thus, management actions that restrict marine and animal dispersal likely have the benefit of limiting the invasive potential of O. drummondii. Keywords coastal dunes; endozoochory; Lepus granatensis; Oryctolagus cuniculus; Spain; thalassochory Acknowledgements We would like to thank graduate students Alicia Alfonso and Brenda Mancheño for assistance in the laboratory and field surveys and the Seville University Glasshouse and Herbarium General Services for their collaboration. We thank the staff of the Paraje Natural Marismas del Odiel and the Biotopo Protegido de Iñurritza for facilities to carry out the field work. Finally, we would like to express our gratitude to Keith 2 MacMillan for translating the manuscript. JGGF collaborated in part of the field and lab work during his sabbatical at the Universidad de Sevilla, Spain (grant CONACYT 7799). This research was supported with MINECO/FEDER, (UE) funds; project CGL2015-65058-R. INTRODUCTION Dispersion is a key process that determines the distribution patterns of species at local, regional and global scales (Auffret et al. 2014). Knowledge of the dispersion mechanisms of species is fundamental to understanding not only the natural processes of colonization and expansion, but also how biological invasions take place (Shigesada and Kawasaki 1997; Higgins and Richardson 1999; Silvertown & Charlesworth 2001). This can make a considerable contribution to early detection and control of invasive species (Puth and Post 2005; Hulme 2015). Biological invasions begin with processes of long-distance dispersal (LDD), brought about either directly or indirectly by human action (Nathan et al 2008). Many different processes facilitate the human-assisted introduction of alien species, but international trade has contributed most to the dispersion of species beyond their natural range (Chapman et al. 2017). The routes of invasion and modes of transport are very diverse and at times complex, but remain little known for many species, especially those that have been unintentionally introduced (Essl et al. 2015). Nevertheless, this information is key to the establishment of effective invasion management strategies (Hulme et al 2008; Chapman et al 2017). Once propagules arrive via LDD, successful establishment of new populations depends on: 1) the availability of suitable habitats, “climate matching” (Sutherst 2003) and “environmental matching” (Ricciardi et al. 2013), 2) the frequency of dispersion events (Drake & Lodge 2006), 3) the capacity of the populations to recolonize the zone following removal by stochastic events (Drury et al. 2007) and 4) the capacity of the species to integrate into the network of interactions of the resident ecosystem through the establishment of mutualist relationships and avoiding or overcoming biotic resistance (Richardson et al. 2000a; Levine et al. 2004; Traveset and Richardson 2014). Introduced plant species that form self-sustainable naturalized populations are potentially invasive (Richardson and Pysek 2012), and transition to the status of invasive species largely depends on the reproductive efficiency and capacity for dispersion of the propagules that act to allow colonization and expansion from the initial points of introduction of the species in the territory (Richardson et al 2000b). The capacity of the propagules of alien plant species to invade natural habitats is greatly enhanced by adaptations to transport by animals, i.e., fleshy fruits (Richardson et al 2000a). However, even in the absence of conspicuous morphological adaptations of fruits or seeds, endozoochory can be an efficient mechanism for expansion of the alien species to new environments (Fernández and Sáiz 2007) even if this is not the manner in which it disperses in its native area. Beach evening-primrose, Oenothera drummondii subsp. drummondii Hook. is a short-lived perennial herb in the family Onagraceae. This plant, which is native to the coastal dunes of the Gulf of Mexico and southeastern USA, is currently distributed in coastal dunes around the world (Dietrich and Wagner, 1988). Its introduction into non-native areas seems to be linked to the movement of goods in which the seeds of 3 O. drummondii were transported as stowaways (Frean et al. 1997; Heyligers 2008). Despite the fact that this species is considered an invasive in some countries or regions, there is little information available about it, for either native or non-native populations. However, recent studies show that, in Mediterranean coastal dunes, O. drummondii presents a higher competitive capacity than the native species (Zunzunegui et al. 2020). Interest in O. drummondii has been raised by the rapid expansion of the species observed in recent decades in some dune systems of southwestern Spain and the high impact it has on the diversity and functioning of the invaded dune systems (García de Lomas et al. 2015; Gallego-Fernández et al. 2019). Comprehensive knowledge of the different mechanisms of dispersion that can be utilized by an invasive species is rarely available, despite the interest in establishing effective measures of control (Gosper et al. 2005). With the objective of establishing a scientific basis for the management of this species, one of the first steps has been to identify the colonization and dispersion mechanisms employed by O. drummondii to maintain its non-native populations and to continue its expansion in zones where it is considered invasive. While O. drummondii is an autochorous species and presents no evident adaptations for long-distance dispersion, two observations made by the authors suggest that this species could present various methods of dispersion: 1) in its native area of the Gulf of Mexico, the appearance of new populations has been observed in sectors of the coast in which the species was not previously recorded, which may suggest that the seeds reached these sectors via seawater dispersion, a mechanism characteristic of many species of beaches and dunes, and 2) partially consumed fruits have been observed in some non-native populations of Spain, suggesting possible dispersion by endozoochory. On the coasts of Spain, The Iberian hare and European rabbit are herbivores that disperse the seeds of a large number of native and non-native plants (Malo and Suárez 1995; Izhaki and Ne’eman 1997; Fernández and Sáiz 2007; Dellafiore et al. 2006, 2010), and thus they could be dispersers of O. drummondii seeds. The purpose of this study was to understand the dispersal modes of O. drummondii in the introduced populations on the coastal dunes of Spain and to determine if they contribute significantly to the invasion process. First, we examined the available records of the distribution of the species on Spanish coastal dunes, which allowed us to reconstruct the history of its spread in the country. Afterwards, we tested whether dispersion by marine currents can be a mechanism of colonization of new areas by O. drummondii, as well as for the species recolonize areas after vegetation loss caused by the impact of storms. Finally, we studied how the species disperses in dune systems, and whether native vertebrate act as dispersers and promote the local expansion of the invasive plant. This information could help to explain the expansion of this species in the Gulf of Cadiz in Spain, and other coasts of the world, and to identify effective control methods in sites where it is considered invasive. METHODS The study species Oenothera drummondii presents actinomorphic-herkogamous yellow flowers and a mixed mating system (self- and cross-pollination) with a high degree of self-compatibility both within its native distribution range and in non-native populations (Gallego-Fernández and García-Franco submitted). It is pollinated by hawkmoths in its native habitats (Wagner et al. 2007). The fruit is a dehiscent tetragonal capsule of 2-3 mm 4 in width and 2.5-5 cm in length that splits longitudinally when wet, releasing numerous small seeds (ca. 0.6 x 1.2 mm) with no evident morphological adaptation for dispersal (Figure 1). In its native range, O. drummondii is relatively scarce and presents small populations (Moreno-Casasola, 1988; Gallego- Fernández and Martínez 2011). The authors have collected seeds from 17 populations along the Gulf of Mexico coast and have found no evidence of predation and/or biotic dispersion (by vertebrates). The species is considered invasive in Spain, Israel and China (Dufour-Dror 2012; Xu et al. 2012; Gallego-Fernández et al. 2019), but has also been introduced into Australia, South Africa, Egypt and New Zealand (ALA 2014; Shaltout et al. 2016; Frean et al. 1997; Heenan et al. 2002). The species has recorded near Spain, in Tetouan, Morocco, although it is not currently present (Hammada et al 2011), and in Marseillan, French Mediterranean coast, where the species has been present since 2014 (Tela Botanica, https://www.telabotanica.org/). To date, however, studies have only been conducted on the effect of its invasion in the dunes of southern Spain, where it has caused considerable modification of the composition and abundance of species of the native plant community (Gallego-Fernández et al. 2019; Zunzunegui et al 2020). The seeds of O. drummondii do not disperse far from plants. The capsules, once mature, open when environmental humidity is high and release the seeds, which fall to the ground beneath the mother plant (unpublished studies of JB Gallego-Fernández et al. of the seed bank of the species show that the majority of the dispersed seeds are found below the mother plant). This type of dispersion strategy is known as ombrohydrochory and is frequently found among other species of the genus Oenothera (Walck and Hidayati 2007). Secondarily, the small sized seeds can be dispersed by the wind to a distance of some metres from the parent (J.B. Gallego-Fernández, personal observation). This type of dispersion is easily observed in the foredune, where the incidence of wind is greater than in the dune zones further inland. At the end of winter, numerous seedlings are observed to the leeward of the parent individuals, extending up to 3-4 m in distance from the mother plant, although the seeds can be dispersed for greater distances (J.B. Gallego-Fernández, personal observation). Species distribution in Spain To determine the dates and locations of the first records of O. drummondii in Spain, we used data collected from the Global Biodiversity Information Facility (GBIF, http://www.gbif.org/), the Spanish Plants Information System (ANTHOS, http://www.anthos.es/), and the Herbarium of the University of Seville. Additional occurrence records for Spain were provided by the Regional Government of Andalusia and collected by field survey along the coast of Gulf of Cádiz and Cantabrian Sea. Subsequent visits were made to all of the dune systems in which the species has been recorded (Figure 1, Table 1). In each site, the population size was recorded, as well as the naturalized or invasive character of the population from the existing literature or through field observation. Populations were considered to be invasive when we observed a high number of individuals, widely distributed in the different habitats of the dune system (Blackburn et al 2011). In addition, the existence of any measure of control of the species and whether the dune system is found within a protected area were also recorded. Moreover, partial or total depredation of the fruits was registered, since this is an indication of the presence of potential seed dispersers. During 5 previous studies (Gallego-Fernández et al 2019), depredated fruits were observed along with the presence of O. drummondii seedlings in the faeces of the Iberian hare (Lepus granatensis Rosenhauer) and European rabbit (Oryctolagus cunniculus L.). For this reason, evidence of traces of the hare and the rabbit in the study sites indicated the presence of possible dispersers. Fig. 1 Location of Oenothera drummondii subsp. drummondii Hook. in Spain based on Herbarium records and recent observations. Dates refer to the first record at that location. The area of the first introduction of the species in the Gulf of Cádiz appears as a grey oval. Grey circles denote the populations recorded, while black circles indicate the populations from which samples and/or fruits and faecal pellets were collected. Dispersal by thalassochory Seed survival in seawater Germination tests were performed to determine whether seeds are capable of survival and germinate following immersion in seawater. The seeds used in this experiment were sourced from undamaged fruits collected in September 2015 in the Dique dunes. Previous germination experiments conducted with seeds from this locality and seeds from other populations of the Gulf of Cadiz showed similar germination patterns (results not shown in this study). The experiment consisted of eight treatments, placing the seeds in containers with 100 ml of seawater for 1, 3, 5, 7, 15, 30 and 60 days. In the control treatment the seeds were submerged in distilled water for 1 day, so that the seeds would have the same hydration as 1 day in seawater at the beginning of the germination experiment. The seawater used came from the same zone and 6 had a salinity (total dissolved solids) of 31.3 ppt at 21 °C. Each immersion treatment included four batches (replicates) of 50 seeds each. The containers were maintained in the laboratory at ambient temperature and the water changed (where applicable) every seven days. Following the immersion period, the seeds were rinsed with deionised water. The lots of seeds were placed in Petri dishes of 90 mm in diameter on a double layer of filter paper moistened with 5 ml of water, and the dishes wrapped with Parafilm. These were left to germinate in a germination chamber with a regime of 12 h of light at a temperature of 20 ºC and 12 h of darkness at 15 ºC for 30 days. Seeds were inspected daily and those that had germinated were counted and removed. Seed germination was considered to have occurred when the radicle became visible. When the seeds were in the seawater, it was observed that the majority sank while only a small proportion floated. Since buoyancy is very important for seed dispersion by marine currents (Guja et al. 2010), an additional experiment was conducted in order to determine the percentage of seeds that can float and their capacity for subsequent germination. Twenty lots of 100 seeds were placed in individual containers with 100 ml of seawater obtained in the Dique’s zone. Every two days, for a period of 8 days, the number of seeds that floated after agitating the sample for 30 s was recorded. After the immersion period, the seeds were separated according to whether or not they floated, and all were rinsed with deionised water. Four replicates of 30 seeds that floated and four replicates of 30 seeds that did not float were allowed to germinate for 30 days, following the same protocol used in the previous germination experiments. Salinity level effect and seed recovery In order to determine the effect of different grades of salinity on seed germination, lots of 100 seeds were placed in Petri dishes on a double layer of filter paper and irrigated with 6 ml of water with eight salinity levels: 0 (control), 50, 100, 200, 300, 400, 500 and 600 mM NaCl. Four replicates per treatment were placed into a germination chamber for 30 days following the same protocol used in the previous germination experiments. A recovery experiment was then conducted in order to confirm the survival of seeds that did not germinate following initial incubation in NaCl. All of these seeds were rinsed in 100 ml deionised water and immediately transferred to Petri dishes lined with filter papers and irrigated with 5 ml deionised water. These seed germination assays were maintained for another 30 days under the same conditions as described above. Dispersal by endozoochory Seed production and fruit predation The production of seeds and intensity of predation of the fruits of O. drummondii were evaluated in six locations that currently have populations of the species (Figure 1). In each site, individuals of similar mean size were selected at random (ca. 40 cm in diameter and 40 cm in height, although the individuals of Playa Baterías were of smaller size, at ca. 20-30 cm in diameter and 30 cm in height) at the end of September, before the first rains of autumn. In each plant selected, all of the mature fruits were counted, recording those that had undergone depredation. We obtained the average of fruit predation per plant for each study 7 population. The mean size of the fruits (length) and mean number of seeds per fruit (seed set) were obtained in each population from 50 randomly selected fruits. Seeds in faecal pellets - collection and germination experiments Based on the visits and recordings of the possible dispersers, the populations from Dique (hares) and Tres Piedras (rabbits) were selected in order to study local scale dispersion. Fresh faecal pellets from hares and rabbits were collected in September just before the rainy season, when predation of hares and rabbits on ripe fruit is highest. The spatial pattern of defecation of the two species is different; rabbits usually defecate in latrines (Delives-Mateos et al. 2008), while hare faeces are scattered (Carro and Soriguer 2010). For this reason, the faeces were collected in different ways. In Dique, 964 hare faecal pellets were collected at random in an area of 2 ha within the semi-stabilized interior dunes, where the density of O. drummondii was high at ca. 6-8 plants/m2. In the dunes of Tres Piedras, the rabbit faecal pellets were collected from 50 latrines (10-20 faeces from each, n=863), distributed in an area of 2.5 ha in the active dunes, where the density of O. drummondii was 0.2-2 plants/m2. The faecal pellets were air-dried in the laboratory and stored in paper envelopes until subsequent processing. A subsample (n = 100) of randomly chosen faecal pellets was taken from those collected at each site and these pellets were individually examined for O. drummondii seeds using a binocular magnifier. The seeds of O. drummondii are easily detected due to their orange-brown colour, small size (ca. 1.2 x 0.6 mm) and elliptical shape, which contrast with the detritus contained in the faecal material. Due to the low frequency of appearance of the seeds in rabbit faecal pellets, the number of pellets examined was increased (to 800 pellets). The average number of seeds per rabbit and hare faecal pellet was calculated. Germination experiments were conducted using the seeds obtained from the hare and rabbit faeces, as well as seeds obtained manually from undamaged mature fruits of each site. Germination experiments were performed in February-March of the following year, with four treatments: seeds from fruits and seeds from hare faecal pellets collected in Dique, and seeds from fruits and seeds from rabbit faecal pellets collected in Tres Piedras. Four replicates of 50 seeds of each treatment were allowed to germinate for a period of 30 days, following the same protocol used in the previous germination experiments. Finally, considering that the hare faecal pellets are often on the beach and can be moved by the sea and marine currents, taking with them the O. drummondii seeds they contain, experiments of survival and seed germination following some time in the seawater were also conducted, following the same protocol used with the seeds derived from fruits. Data analyses All data were tested for normality with the Kolmogorov-Smirnov and Lilliefors methods prior to analysis. Due to their non-normal distributions, the differences in the number of fruits per plant, fruit size, number 8 of seeds per fruit, number of predated fruits per plant, proportion of faeces with seeds and time to first germination of the salinity treatments were compared between sites using Kruskal–Wallis H tests and Mann–Whitney U tests. A set of ANOVAs were used to detect differences in the final accumulated percentage of germination within the different sets of treatments, and a t-student test was used to detect differences in the proportion of plants with signs of depredation among sampling sites. An arc-cosine transformation was applied to the final accumulated percentage of germination data before performing parametric statistical analyses. A step-down pair-wise posteriori comparison was based on the Ryan-Einot- Gabriel-Welsh F (REGWF) procedure based on the F test (Meyers et al 2017). All statistical analyses were performed with SPSS 25.0 (SPSS Corp., Chicago, Illinois, USA). RESULTS Introduction and expansion in Spain The first record in northern Spain is from the year 1915 in Cestona, an inland location at 8 km from the coast (Silvestre 1980) (Figure 1; Table 1). There are no references regarding how the species arrived. The lack of subsequent records indicates that the abundance and distribution of the species in the coastal region had not been considered very important. It is not until 1961 that it is cited once again, this time in the coastal dune system of Zarautz (Aizpuru et al. 1996). The species was also recorded in 1982 on Orio beach (Aizpuru et al. 1996) and on the Zumaia dunes in 2002, both zones very close to Zarautz, at 3 and 7 km, respectively. Intensive samples of the coastal flora conducted in 2001 (Silvan and Campos 2002), and field samples taken in 2017 by the authors, indicated that the species is currently only found in the Zarautz dunes. In 2017, control of the population began in the form of manual removal. Table 1. Sites in Spain where the presence of Oenothera drummondii has been recorded. For each site, the geographic location, date of first record, presence of possible disperser animals, population size, current status of the population, history of control actions and location of the population within a protected area were recorded Location Latitude Longitude First record Potential Seed dispersers Actual Population size Current Status Control management Protected area North Cestona 43°14’N; 02°15’W 1915 - - - - No Zarautz 43°17’N; 02°09’W 1961 No ≈ 300 Naturalize Yes Yes Orio 43°17’N; 02°07’W 1982 No - Does not exist No No Zumaia 43°18’N; 02°14’W 2002 No - Does not exist No No South west Playa Baterías 36°39’N; 06°24’W 1957 No 76 Naturalized No No Tres Piedras 36°42’N, 06°25’W 1978 Rabbits >10,000 Invasive Yes No Doñana 36°49´N, 06°24´W 1978 Hares, rabbits, red and fallow deers 6 Eradicated 2015 Yes Yes 9 La Barrosa 36°21’N, 06°10’W 1980 No 21 Naturalize Yes No Dique 37°09’N; 06°54’W 1996 Hares >4 Millions Invasive Yes Yes Mazagón 37°07’N; 06°48’W 2007 No - Does not exist No No Asperillo 37°04’N; 06°41’W 2012 No 1 Eradicated 2012 Yes Yes Punta Umbría 37°10’N; 06°56’W 2013 Rabbits 4 Eradicated 2017 Yes No In southern Spain (Figure 1), the first record was in the year 1957 around the village of Rota (Silvestre 1980). Since then, it has been recorded in another seven coastal localities along the length of 120 km of the coast of the Gulf of Cádiz. The next records are from 1978 in the dunes of Tres Piedras and Doñana (Silvestre 1980) located at 10 and 25 km in distance along the coast, respectively, to the north of Rota. On the beach of La Barrosa, the species was recorded in 1980 at a site 35 km to the south of Rota. In 1996, O. drummondii was recorded for the first time in the dunes of Dique (García-Mora pers. comm.), located around 80 km from Rota and 55 km from Doñana. Around Dique, some isolated individuals were recorded in 2007 in Mazagón at a distance of 14 km (de las Heras 2007), in 2012 in Asperillo at 20 km (Consejería de Medio Ambiente 2013) and in 2013 in Punta Umbría at 1 km (García de Lomas et al. 2015). Oenothera drummondii control programs involving extraction of the plants, conducted by the regional government, have reduced (La Barrosa) or even eradicated (Doñana, Punta Umbría and Asperillo) its presence (Table 1). In the case of Doñana, located in a National Park, rigorous control has been applied since the end of the 1990s, such that it is currently considered eradicated (Consejería de Medio Ambiente 2016). The last six plants that remained in Doñana were removed and used in this study. However, in the dunes of Dique and Tres Piedras, despite the control actions undertaken, the species is considered to present invasive behaviour (García de Lomas et al. 2015; Gallego-Fernández et al. 2019). The first record of introduction to the Gulf of Cádiz, in Rota, does not provide exact information regarding its location. Currently, there are two small-sized populations of O. drummondii in this municipality; one in a highway ditch situated around 1.5 km from the coast with around 50 individuals and another on the coast at Playa Baterías, with ca. 76 plants situated on the edge of a small sandy cliff next to the beach, with evidence of occasional erosion by the marine swell. Dispersal by thalassochory Seed survival in seawater Seed germination was affected by the duration of immersion in seawater (F6,56=51.86, p<0.0001) and the origin of the seeds (F1,56=113.28, p<0.0001) (Figure 2). The final accumulated germination of seeds from fruits remained relatively high (86-90%) and similar to the control lots until 15 days of immersion in seawater, but it subsequently decreased significantly to 67 ± 4.8% and 13 ± 4.8% after 30 and 60 days of 16 Dispersion of propagules by waves and ocean currents has been described as a key factor in the rapid expansion of species that invade beaches and dunes over long distances (Heyligers 2007; Konlechner et al 2016). Dispersion can be as seeds, such as the case of Cakile marítima and C. edentula over thousands of kilometres of coasts of Australia and California (Barbour and Rodman 1970; Rodman 1986) and Euphorbia paralias on the coasts of Australia (Heyligers 2002), or seeds and fragments of plants, as occurs with Ammophila arenaria on the coasts of California, Australia and New Zealand (Buell et al. 1995; Konlechner and Hilton 2009). The capacity for dispersion by seawater can also maintain invasive populations, enabling them to overcome both natural disturbances and management actions intended to control them (Buell 1995; Hilton et al. 2005; Heyligers 2007; Rodman 1986). On the other hand, the results of our experiments have shown that O. drummondii seeds are tolerant to salinity, even when exposed to concentrations equivalent to that of seawater (600 mM NaCl), and that their germination at relatively low saline concentrations levels is similar to when exposed to non-saline water. This response pattern has been described in other dune species worldwide (Maun 2009). If the seeds of O. drummondii, floating or otherwise, are deposited by the waves on the driftline of the beaches, germination will be impeded by the high salinity. However, since they are of small size, they can be lifted and carried by the wind (Maun 2009) towards the back beach, foredunes and even inland dunes. Moreover, on occasions, a high storm surge may transport seeds inland (Yang et al 2012). In such cases, the seeds can reach “safe sites”, with soils of low salinity, where they can be washed by the rain and begin germination. Endozoochory, the key to invasion This study has demonstrated that the seeds of O. drummondii are dispersed through endozoochory by Iberian Hares and European rabbits. Both of these herbivorous species are recognized as seed-dispersal agents (Pakeman et al. 2003; Dellafiore et al. 2010, Mancilla-Leytón et al. 2013) that can contribute to the expansion of alien species (Traveset and Richardson 2006; Fernández and Sainz 2007; Calvino-Cancela 2011), although the role of the hare has been little studied to date. Iberian Hares and European rabbits are considered foragers, and ingestion of seeds is accidental according to the hypothesis of “the foliage is the fruit” (Janzen 1984), and they thus act as non-specific dispersers of small and dry seeds (Malo et al., 1995; Izhaki and Ne’eman 1997; Fernández and Sáiz 2007; Dellafiore et al. 2010). However, in the case of O. drummondii, these two dispersers act as frugivores, since they consume the fruits of the plant directly (Figure 4), acting as specific dispersers (legitimate dispersers), despite the fruits and seeds of O. drummondii have no specific traits for dispersion by animals, for which reason this is also a case of a non-standard dispersal mechanism (Higgins 2003). Although the predation of fruits of O. drummondii is low (Dique 7.3%, Tres Piedras 16.7%), the effectiveness of the dispersion (Malo and Suárez 1998) is very high, since: i) the quantity of seeds ingested is very high, ii) 63-73% of the seeds that pass through the digestive system of either frugivore germinate, which is a high level of survival of seeds following consumption by generalist herbivores (Mouissie et al. 2004), and iii) the seed dispersion carried out by each disperser is high, although it does differ according to species (31.4 and 0.3 seeds per hare and rabbit faecal pellet, respectively). Indeed, if we consider that each hare produces around 200 faecal pellets per day (Langbein et al 1999) and each rabbit produces 373 faecal 17 pellets per day (González-Redondo 2009), one hare can disperse around 6280 seeds per day (of which, 73% can germinate) while a rabbit disperses around 112 seeds per day (of which, 63% germinate). It should be noted that the only two sites where the species presents invasive behaviour are those in which the disperser animals of their seeds are present: Dique and Tres Piedras. The mutualist interaction established between O. drummondii and the native dispersers must play an important role in the expansion of this alien species, since it provides a complementary mechanism of short-distance dispersal, allowing the plant to colonize new sites, contributing to its local expansion (Nathan and Muller-Landau 2000) and facilitating the process of invasion (Richardson and Pysek 2012). The current distribution of O. drummondii in the coastal zones of Spain following its introduction ca. 60 years ago, is due not only to its great reproductive success, but also to the different mechanisms of seed dispersion it has been able to establish in its sites of colonization, where it presents relationships of frugivory with the local fauna. Conclusions Our study has shown that O. drummondii has been able to establish in several dune systems in the north and southwest of Spain since the middle of the 20th century. Regional expansion from the original point of introduction may have been caused by seed dispersal through ocean currents, as our experiments showed that some of the seeds can float in seawater and maintain the ability to germinate. In most of the dune systems where the species has been recorded, it has been maintained in small naturalized populations, some of which have been eradicated through early and sustained management. However, in the Dique and Tres Piedras populations, the species has an invasive behaviour. This is possibly because here, the seeds of O. drummondii are locally dispersed by endozoochory, by the Iberian hare and the European rabbit. At regional scale, we recommend management measures to prevent the arrival of the species to coastal areas near existing populations. Management efforts should focus on beaches and foredunes to prevent the establishment of germinated storm-dispersed seeds. At a local level, in dunes where the rabbits and hares are present, we recommend management measures to contain the invasive species in the invaded areas by eliminating foci of invasion resulting from endozoochory. Once dispersal processes are known, both regional and local management can be coordinated to control and eradicate O. drummondii in the invaded dune area. References Aizpuru I, Aparicio JM, Aperribay JA, Aseginolaza C, Elorza J, Garin F, Patino S, Pérez-Dacosta JM, Pérez de Ana JM, Uribe PM, Urrutia P, Valencia J, Vivant J (1996) Contribuciones al conocimiento de la flora del País Vasco. An Jard Bot Mad 54:419-435 ALA (2014) The Atlas of Living Australia: http://www.ala.org.au Andersen UV (1993) Dispersal strategies of Danish seashore plants. Ecography 16:289–298. https://doi.org/10.1111/j.1600-0587.1993.tb00218.x 18 Auffret AG, Berg J, Cousins SAO (2014) The geography of human-mediated dispersal. Divers Distrib 20:1450– 1456. https://doi.org/10.1111/ddi.12251 Barbour MG, Rodman JE (1970) Saga of the West Coast sea-rockets: Cakile edentula ssp. californica and C. maritima. Rhodora 72:370- 386. https://www.jstor.org/stable/23311387 Blackburn TM, Pyšek P, Bacher S, Carlton JT, Duncan RP, Jarosík V, Wilson JRU, Richardson DM (2011) A proposed unified framework for biological invasions. Trends Ecol Evol 26: 333–339 https://doi.org/10.1016/j.tree.2011.03.023Buell AC, Pickart AJ, Stuart JD (1995) Introduction history and invasion patterns of Ammophila arenaria on the north coast of California. Conserv Biol 9:1587–1593. https://doi.org/10.1046/j.1523-1739.1995.09061587.x Calvino-Cancela M (2011) Seed dispersal of alien and native plants by vertebrate herbivores. Biol Invasions 13:895-904. https://doi.org/10.1007/s10530-010-9877-6 Carro F, Soriguer RC (2010) La liebre ibérica. Organismo Autónomo de Parques Nacionales. Madrid Chapman D, Purse BV, Roy HE, Bullock JM (2017) Global trade networks determine the distribution of invasive non-native species. Glob Ecol Biogeogr 26:907–917. https://doi.org/10.1111/geb.12599 Colmer T D, Flowers T J (2008) Flooding tolerance in halophytes. New Phytol 179:964–974. https://doi.org/10.1111/j.1469-8137.2008.02483.x Consejería de Medio Ambiente (2013) Espacio Natural de Doñana. Memoria de actividades y resultados 2012. Consejería de Medio Ambiente y ordenación del Territorio, Junta de Andalucía, Sevilla Consejería de Medio Ambiente (2016) Espacio Natural de Doñana. Memoria de actividades y resultados 2015. Consejería de Medio Ambiente y ordenación del Territorio, Junta de Andalucía, Sevilla de las Heras MA. 2007. Estudio de flora y vegetación del oeste del entorno de Doñana. Tesis doctoral. Departamento de Biología Ambiental y Salud Pública, Universidad de Huelva. 540 pp Delibes-Mateos M, Delibes M, Ferreras P, Villafuerte R (2008) Key role of European rabbits in the conservation of the Western Mediterranean basin hotspot. Conserv Biol 22:1106-1117. https://doi.org/10.1111/j.1523- 1739.2008.00993.x Dellafiore C, Muñoz S, Gallego Fernández JB (2006) Rabbits (Oryctolagus cuniculus) as dispersers of Retama monosperma (L.) Bois seeds in a coastal dune system. Ecoscience 13:5-10. https://doi.org/10.2980/1195- 6860(2006)13[5:ROCADO]2.0.CO;2 Dellafiore CM, Gallego-Fernández JB, Muñoz S (2010) The rabbit (Oryctolagus cuniculus) as a seed disperser in a coastal dune system. Plant Ecol 206:251-261. https://doi.org/10.1007/s11258-009-9639-7 Dietrich W, Wagner WL (1988) Systematics of Oenothera section Oenothera subsection Raimannia and subsection Nutantigemma (Onagraceae). Monogr Systematic Bot 24:1-91. https://www.jstor.org/stable/i25027711 19 Drake JM, Lodge DM (2006) Allee effects, propagule pressure and the probability of establishment: risk analysis for biological invasions. Biol Inv 8:365– 375. https://doi.org/10.1007/s10530-004-8122-6 Drury K, Drake J, Lodge D, Dwyer G (2007) Immigration events dispersed in space and time: factors affecting invasion success. Ecol Modell 206:63–78. https://doi.org/10.1016/j.ecolmodel.2007.03.017 Dufour-Dror JM (2012) Alien invasive plants in Israel. The Middle East Nature Conservation Promotion Association, Ahva, Jerusalem Essl F, Bacher S, Blackburn T, Booy O, Brundu G, Brunel S, et al. (2015) Crossing frontiers in tackling pathways of biological invasions. BioScience 65:769–782. https://doi.org/10.1093/biosci/biv082 Fernández A, Sáiz F (2007) The European rabbit (Oryctolagus cuniculus L.) as seed disperser of the invasive opium poppy (Papaver somniferum L.) in Robinson Crusoe Island, Chile. Mastozool Neotrop 14:19–27. http://www.scielo.org.ar/scielo.php?script=sci_arttext&pid=S0327-93832007000100003 Frean M, Balkwill K, Goid C, Burt S (1997) The expanding distributions and invasiveness of Oenothera in southern Africa. S Afr J Bot 63:449-458. https://doi.org/10.1016/S0254-6299(15)30798-59-458 Gallego-Fernández JB, Martínez ML (2011) Environmental filtering and plant functional types on Mexican foredunes along the Gulf of Mexico. Ecoscience 18:52-62. https://doi.org/10.2980/18-1-3376 Gallego-Fernández JB, Martínez ML, García-Franco JG, Zunzunegui M (2019) The impact on plant communities of an invasive alien herb, Oenothera drummondii, varies along the beach-coastal dunes gradient. Flora 260, 151466. https://doi.org/10.1016/j.flora.2019.151466 Gallego-Fernández JB, Morales-Sanchez JA, Martínez ML, García-Franco JG, Zunzunegui M (2020) Recovery of beach-foredune vegetation after disturbance by storms. J Coast Res 95(sp1):34-38. https://doi.org/10.2112/SI95-007.1 García-de-Lomas J, Fernández-Carrillo L, Saavedra M, Mangas L, Rodríguez-Hiraldo C, Martínez-Montes E (2015) Invasión de Oenothera drummondii Hook. (Onagraceae) en el Paraje natural Marismas del Odiel (Huelva, S España): bases para la gestión de una invasión avanzada. Rev Soc Gad Hist Nat 9:41-50 González‐Redondo P (2009) Number of faecal pellets dropped daily by the wild rabbit (Oryctolagus cuniculus). J Anim Vet Adv 8:2635–2637. http://hdl.handle.net/11441/16210 Gosper CR, Stansbury CD, Vivian-Smith G (2005) Seed dispersal of fleshy-fruited invasive plants by birds: contributing factors and management options. Divers Distrib 11:549–558. https://doi.org/10.1111/j.1366- 9516.2005.00195.x Guja LK, Merritt DJ, Dixon KW (2010) Buoyancy, salt tolerance and germination of coastal seeds: implications for oceanic hydrochorous dispersal. Funct Plant Biol 37:1175–1186. https://doi.org/10.1071/FP10166 20 Guja LK, Merritt DJ, Dixon KW, Wardell-Johnson G (2014) Dispersal potential of Scaevola crassifolia (Goodeniaceae) is influenced by intraspecific variation in fruit morphology along a latitudinal environmental gradient. Australian J Bot 62: 56–64. https://doi.org/10.1071/BT13290 Hammada S, Linares L, Cortes J (2011) Biodiversité floristique des dunes littorales de l’Oued El Maleh (Martil) et du bas Tahaddart: résultats préliminaires. In: Bayed A (ed) Sandy beaches and coastal zone management – Proceedings of the Fifth International Symposium on Sandy Beaches, 19th-23rd October 2009, Rabat, Morocco. Travaux de l'Institut Scientifique, Rabat, série générale, 2011, n°6, 45-50. Heenan PB, de Lange PJ, Cameron EK, Champion PD (2002) Checklist of dicotyledons, gymnosperms, and pteridophytes naturalised or casual in New Zealand: Additional records 1999–2000. New Zeal J Bot 40:155-174. https://doi.org/10.1080/0028825X.2002.9512780 Hesp P (1991) Ecological processes and plant adaptations on coastal dunes. J Arid Environ 21:165-191. https://doi.org/10.1016/S0140-1963(18)30681-5 Heyligers PC (2002) The spread of the introduced Euphorbia paralias (Euphorbiaceae) along the mainland coast of south-eastern Australia. Cunninghamia 7:563–578 Heyligers PC (2007) The role of currents in the dispersal of introduced seashore plants around Australia. Cunninghamia 10:167–188 Heyligers PC (2008) Flora of the Stockton and Port Hunter sandy foreshores with comments on fifteen notable introduced species. Cunninghamia 10:493-511 Higgins SI, Nathan R, Cain ML (2003) Are long‐distance dispersal events in plants usually caused by nonstandard means of dispersal? Ecology 84:1945–1956. https://doi.org/10.1890/01-0616 Higgins SI, Richardson DM (1999) Predicting Plant Migration Rates in a Changing World: The role of Long- Distance Dispersal. Am Nat 153:464-475. https://doi.org/10.1086/303193 Hilton MJ, Duncan M, Jul A (2005) Processes of Ammophila arenaria (marram grass) invasion and indigenous species displacement, Stewart Island, New Zealand. J Coast Res 21:175–185. https://doi.org/10.2112/01041.1 Hulme PE (2015) Invasion pathways at a crossroad: policy and research challenges for managing alien species introductions. J Appl Ecol 52:1418–1424. https://doi.org/10.1111/1365-2664.12470 Hulme PE, Bacher S, Kenis M, Klotz S, Kuhn I, Minchin D, et al. (2008) Grasping at the routes of biological invasions: a framework for integrating pathways into policy. J Appl Ecol 45:403–414. https://doi.org/10.1111/j.1365-2664.2007.01442.x Izhaki I, Ne‘eman G (1997) Hares (Lepus spp.) as seed dispersers of Retama raetam (Fabaceae) in a sandy landscape. J Arid Environ 37:343–354. https://doi.org/10.1006/jare.1997.0273 21 Janzen DH (1984) Dispersal of small seeds by big herbivores: foliage is fruit. Am Nat 123:338–353. https://doi.org/10.1086/284208 Konlechner TM, Orlovich DA, Hilton MJ (2016) Restrictions in the sprouting ability of an invasive coastal plant, Ammophila arenaria, from fragmented rhizomes. Plant Ecol 217:521–532. https://doi.org/10.1007/s11258- 016-0597-6 Konlechner TM, Hilton MJ (2009) The Potential for Marine Dispersal of Ammophila Arenaria (Marram Grass) Rhizome in New Zealand. J Coast Res SI56:434-437. https://www.jstor.org/stable/25737613 Langbein J, Hutchings MR, Harris S, Stoate C, Tapper SC, Wray S (1999) Techniques for assessing the abundance of Brown Hares Lepus europaeus. Mamm Rev 29:93-116. https://doi.org/10.1046/j.1365- 2907.1999.00040.x Levine JM, Adler PB, Yelenik SG (2004). A meta‐analysis of biotic resistance to exotic plant invasions. Ecol Lett 7:975–989. https://doi.org/10.1111/j.1461-0248.2004.00657.x Malo JE, Suárez F (1995) Establishment of pasture species on cattle dung: the role of endozoochorous seeds. J Veg Sci 6:169–174. https://doi.org/10.2307/3236211 Malo JE, Suarez F (1998) The dispersal of dry-fruited shrub by reed deer in a Mediterranean ecosystem. Ecography 21:204-211. https://doi.org/10.1111/j.1600-0587.1998.tb00673.x Mancilla-Leytón JM, González-Redondo P, Martín Vicente A (2013) Effects of rabbit gut passage on seed retrieval and germination of three shrub species. Basic Appl Ecol 14:585-592. https://doi.org/10.1016/j.baae.2013.08.005 Maun MA (2009) The Biology of Coastal Sand Dunes. Oxford University Press, New York, NY Meyers L S, Gamst G, Guarino AJ (2017) Applied Multivariate Research: Design and Interpretation. SAGE Publications Singapore. Moreno-Casasola P (1988) Patterns of plant species distribution on Mexican coastal dunes along the Gulf of Mexico. J Biogeogr 15:787–806. https://doi.org/10.2307/2845340 Mouissie AM, van der Veen E J, Veen GF, van Diggelen R (2005) Ecological correlates of seed survival after ingestion by fallow deer. Funct Ecol 19:284-290. https://doi.org/10.1111/j.0269-8463.2005.00955.x Nathan R, Muller‐Landau H C (2000) Spatial patterns of seed dispersal, their determinants and consequences for recruitment. Trends Ecol Evol 15:278– 285. https://doi.org/10.1016/S0169-5347(00)01874-7 Nathan R, Schurr FM, Spiegel O, Steinitz O, Trakhtenbrot A., Tsoar A (2008) Mechanisms of long‐distance seed dispersal. Trends Ecol Evol 23:638–647. https://doi.org/10.1016/j.tree.2008.08.003 Pakeman, R. J., Digneffe, G., & Small, J. L. (2002). Ecological correlates of endozoochory by herbivores. Funct Ecol 16: 296–304. https://doi.org/10.1046/j.1365-2435.2002.00625.x 22 Puth LM, Post DM (2005) Studying invasion: have we missed the boat? Ecol Lett 8:715-721. https://doi.org/10.1111/j.1461-0248.2005.00774.x Ricciardi A, Hoopes MF, Marchetti MP, Lockwood JL (2013) Progress toward understanding the ecological impacts of nonnative species. Ecol Monogr 83:263– 282. https://doi.org/10.1890/13-0183.1 Richardson DM, Allsopp N, D'Antonio CM, Milton SJ, Rejmanek M (2000a) Plant invasions - the role of mutualisms. Biol Rev Cambridge Philosophic Soc 75:65-93. https://doi.org/10.1111/j.1469- 185X.1999.tb00041.x Richardson DM, Pysek P, Rejmánek M, Barbour MG, Panetta FD, West CJ (2000b) Naturalization and invasion of alien plants: concepts and definitions. Divers Distrib 6:93–107. https://doi.org/10.1046/j.1472- 4642.2000.00083.x Richardson, DM, Pyšek, P (2012) Naturalization of introduced plants: ecological drivers of biogeographical patterns. New Phytol 196:383–396. https://doi.org/10.1111/j.1469-8137.2012.04292.x Ridley HN (1930) The dispersal of plants throughout the world. L Reeve and Co., Kent Rodman J (1986) Introduction, establishment and replacement of sea-rockets (Cakile, Cruciferae) in Australia. J Biogeogr 13:159-171. https://doi.org/10.2307/2844990 Shaltout KH, Hosni HA, El-Kady HF, El-Beheiry MA, Shaltout SK (2016) Composition and pattern of alien species in the Egyptian flora. Flora 222:104–110. https://doi.org/10.1016/j.flora.2016.04.001 Shigesada N, Kawasaki K (1997) Biological Invasions: Theory and Practice, Oxford University Press Silván F, Campos JA (2002) Estudio de la flora vascular amenazada de los arenales la Comunidad Autónoma del País Vasco. Departamento de Agricultura y Pesca del Gobierno Vasco, Vitoria. Silvertown, J, Charlesworth D (2001) Introduction to Plant Population Biology. Blackwell, London Silvestre S (1980) Notas breves 15. Oenothera drummondii Hook. Lagascalia 9:244-245 Sutherst RW (2003) Prediction of species geographical ranges. J Biogeogr 30:805– 816. https://doi.org/10.1046/j.1365-2699.2003.00861.x Traveset A, Richardson DM (2006) Biological invasions as disruptors of plant reproductive mutualisms. Trends Ecol Evol 21:208–16. https://doi.org/10.1016/j.tree.2006.01.006 Traveset A, Richardson DM (2014) Mutualistic interactions and biological invasions. Annu Rev Ecol Evol Syst 45:89–113. https://doi.org/10.1146/annurev-ecolsys-120213-091857 Wagner WL, Hoch PC, Raven PH (eds) (2007) Revised classification of the Onagraceae. Syst Bot Monogr 83:1– 240 23 Walck JL, Hidayati SN (2007) Ombrohydrochory and its relationship to seed dispersal and germination strategies in two temperate North American Oenothera species (Onagraceae). Int J Plant Sci 168:1279–1290. https://doi.org/10.1086/521691 Xu H, Qiang S, Genovesi P, Ding H, Wu J, Meng L, Han Z, Miao J, Hu B, Guo J, Sun H, Huang C, Lei J, Le Z, Zhang X, He S, Wu Y, Zheng Z, Chen L, Jarošík V, Pyšek P (2012) An inventory of invasive alien species in China. NeoBiota 15:1-26. https://doi.org/10.3897/neobiota.15.3575 Yang H, Lu Q, Wu B, Zhang J (2012) Seed dispersal of east Asian coastal dune plants via seawater – short and long distance dispersal. Flora 207:701–706. https://doi.org/10.1016/j.flora.2012.08.001 Zunzunegui M, Ruiz-Valdepeñas E, Sert MA, Díaz-Barradas MC, Gallego-Fernández JB (2020) Field comparison of ecophysiological traits between an invader and a native species in a Mediterranean coastal dune. Plant Physiol Biochem 146:278-286. https://doi.org/10.1016/j.plaphy.2019.11.032